Keeping first vs last vs none in Pandas - Practice Questions

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of drop_duplicates with keep='first'

What is the output DataFrame after running this code?

Pandas

import pandas as pd

df = pd.DataFrame({
    'A': [1, 2, 2, 3, 3, 3],
    'B': ['x', 'y', 'y', 'z', 'z', 'z']
})

result = df.drop_duplicates(subset=['A'], keep='first')
print(result)

A

   A  B
1  2  y
3  3  z

B

   A  B
2  2  y
4  3  z

C

   A  B
0  1  x
2  2  y
5  3  z

D

   A  B
0  1  x
1  2  y
3  3  z

💡 Hint

Keep='first' keeps the first occurrence of each duplicate group.

drop_duplicates with keep='first' keeps the first row for each unique value in column 'A'. Rows with A=2 and A=3 appear multiple times, but only the first occurrence is kept.

❓ Predict Output

intermediate

Output of drop_duplicates with keep='last'

What is the output DataFrame after running this code?

Pandas

import pandas as pd

df = pd.DataFrame({
    'A': [1, 2, 2, 3, 3, 3],
    'B': ['x', 'y', 'y', 'z', 'z', 'z']
})

result = df.drop_duplicates(subset=['A'], keep='last')
print(result)

A

   A  B
0  1  x
1  2  y
3  3  z

B

   A  B
1  2  y
4  3  z

C

   A  B
0  1  x
2  2  y
5  3  z

D

   A  B
2  2  y
4  3  z

💡 Hint

Keep='last' keeps the last occurrence of each duplicate group.

drop_duplicates with keep='last' keeps the last row for each unique value in column 'A'. For A=2, last occurrence is index 2; for A=3, last occurrence is index 5.

❓ Predict Output

advanced

Output of drop_duplicates with keep=False

What is the output DataFrame after running this code?

Pandas

import pandas as pd

df = pd.DataFrame({
    'A': [1, 2, 2, 3, 3, 3],
    'B': ['x', 'y', 'y', 'z', 'z', 'z']
})

result = df.drop_duplicates(subset=['A'], keep=False)
print(result)

A

   A  B
1  2  y
3  3  z

B

   A  B
0  1  x

C Empty DataFrame\nColumns: [A, B]\nIndex: []

D

   A  B
0  1  x
1  2  y
3  3  z

💡 Hint

Keep=False drops all duplicates, keeping only unique rows.

drop_duplicates with keep=False removes all rows that have duplicates in column 'A'. Only A=1 is unique, so only that row remains.

🔧 Debug

advanced

Identify the error in drop_duplicates usage

What error does this code raise when executed?

Pandas

import pandas as pd

df = pd.DataFrame({
    'A': [1, 2, 2],
    'B': ['x', 'y', 'y']
})

result = df.drop_duplicates(subset=['A'], keep='middle')
print(result)

A ValueError: keep must be either 'first', 'last', or False

B TypeError: unhashable type: 'list'

C KeyError: 'middle'

D No error, outputs DataFrame with all rows

💡 Hint

Check the allowed values for the keep parameter.

The keep parameter only accepts 'first', 'last', or False. Using 'middle' raises a ValueError.

🚀 Application

expert

Count unique rows after drop_duplicates with different keep values

Given this DataFrame, how many rows remain after applying drop_duplicates on column 'A' with keep='first', keep='last', and keep=False respectively?

Pandas

import pandas as pd

df = pd.DataFrame({
    'A': [1, 2, 2, 3, 3, 3, 4],
    'B': ['x', 'y', 'y', 'z', 'z', 'z', 'w']
})

count_first = len(df.drop_duplicates(subset=['A'], keep='first'))
count_last = len(df.drop_duplicates(subset=['A'], keep='last'))
count_none = len(df.drop_duplicates(subset=['A'], keep=False))

print(count_first, count_last, count_none)

A 4 4 2

B 4 4 1

C 5 5 3

D 5 5 2

💡 Hint

Count unique values and consider which rows are kept or removed.

There are 4 unique values in 'A': 1, 2, 3, 4. keep='first' and keep='last' keep one row per unique value, so 4 rows remain. keep=False removes all duplicates, so only unique values that appear once remain: 1 and 4, so 2 rows remain.